Predictive Accuracy of Serial Transvaginal Cervical Lengths and Quantitative Vaginal Fetal Fibronectin Levels for Spontaneous Preterm Birth Among Nulliparous Women
Prediction of spontaneous preterm birth, a leading cause of infant mortality, depends in practice largely on a history of previous birth, and the risk for first time mothers and nulliparous women is more difficult to assess. The aim of this study was to assess the reliability of serial transvaginal cervical length (CL) and quantitative fetal fibronectin (fFN) measurements in predicting spontaneous preterm birth (before 37 weeks) in a large, prospective cohort of nulliparous women.
The accuracy of screening was tested through an observational study of 9410 nulliparous women (demographics: median age, 27.0 years [interquartile range, 9.0 years]; race, 60.7% non-Hispanic white, 13.8% non-Hispanic black, 16.5% Hispanic, 4.0% Asian, and 5.1% other) with singleton pregnancies carried to 20 weeks or more from 8 US clinical sites between October 2010 and May 2014. Transvaginal CL and quantitative vaginal fFN levels were measured and assessed at 2 study visits a month or more apart. Spontaneous preterm births at less than 37 weeks and less than 32 weeks were the primary and secondary outcomes. Of the 9410 women, 91.4% had term births (8601), whereas medically indicated preterm births occurred in 3.6% of cases (335 women). Among the women who had spontaneous preterm birth (474 or 5.0%), CL of 25 mm or less occurred in 8% (35 of 439) at 16 to 22 weeks' gestation and in 23.3% (94 of 403) at 22 to 30 weeks' gestation. In contrast, a CL of less than 25 mm was found in approximately 2% of women with a term birth, and the positive predictive value of CL was 16%. Fetal fibronectin levels of 50 ng/mL or greater were present in 7.3% of cases (30/410) at 16 to 22 weeks and 8.1% (31/384) at 22 to 30 weeks in which spontaneous preterm births occurred. The area under the receiver operating characteristic (AUC) curve at 22 to 30 weeks for fFN of 50 ng/mL or greater was 0.59 (95% confidence interval, 0.56-0.62) and for transvaginal CL of 25 mm or less was 0.67 (95% confidence interval, 0.64-0.70); the AUC for the combination was the same as for transvaginal CL.
Fetal fibronectin and serial transvaginal ultrasound measurement of CL, alone or in combination, were found to have low accuracy for the prediction of preterm birth in nulliparous women with singleton pregnancies. Therefore, the routinization of this screening is not recommended.